AI governance and security consulting pricing at DSE runs from about $1,500 for a small-business security posture check to $300,000 or more for a full implementation or private AI build, priced as named engagement models rather than one list price. Every figure is a non-binding market-estimate range, and the final fee is fixed in writing after a scoping call, never billed hourly.
We price work in named engagement models: governance and readiness, AI security, implementation, data, and private AI. Each model carries a non-binding market estimate range, not a fixed list price. We scope and fix every engagement in writing after a call. The range below is where the conversation starts, not where it ends.
All figures are non-binding market estimate ranges, not quotes. Final fees are fixed in writing after a scoping call. Readiness and assessment work is not certification. It does not guarantee any regulatory or compliance outcome. Governance frameworks are informed by NIST AI RMF and designed to support your EU AI Act compliance program. They are aligned to those principles, not a substitute for your counsel or an accredited audit.
Each model below carries the range we see in the market for work of that shape. Complex programs are scoped separately. Multi-entity, cross-jurisdiction, or enterprise-wide work lands toward, or above, the top of a range. Pick the model that sounds like your problem. Or run the estimate tool below.
You get AI inventory, risk tiering, and human-review and vendor controls. We settle the vendor-versus-private-boundary decision. Then a short remediation roadmap, before rollout expands.
Scope the healthcare snapshot →You get AI use-case inventory, NIST AI RMF risk classification, and system and tool-boundary decisions. Plus attributable logging requirements and a short readiness roadmap. Program, security, and procurement teams can all review it.
Scope the federal brief →You get AI inventory, control crosswalk, and gap findings. The governance program aligns to the supervisory expectations your institution already answers to.
Scope the finserv assessment →You get deployment-boundary options, access and logging design, and the vendor-versus-private-path decision. Plus the operating evidence a regulated team needs before committing to a platform path.
Scope the architecture brief →Qualification note: these ranges assume a single business unit, one program, or one in-scope workflow. Bigger work scopes separately and lands higher. That includes multi-entity healthcare systems, enterprise-wide governance builds, federal ATO/FedRAMP-style paths, and large private-AI platform programs.
A fixed-scope read on data, governance, infrastructure, and talent maturity. Plus a shadow-AI audit and a prioritized roadmap. You fund the right build first.
Scope a readiness sprint →You get AI inventory, control crosswalk, and gap findings. The governance program aligns to NIST AI RMF principles. You also get a prioritized remediation roadmap.
Scope a gap assessment →A DPIA and privacy impact assessment for an AI system. It follows GDPR Article 35 structure and US state risk-assessment format. You get a jurisdiction trigger analysis, a risk register, and a residual-risk statement. Readiness and advisory, not legal advice.
Scope a DPIA assessment →An exposure map across wire, treasury, help-desk, executive-support, and vendor-payment flows. A review of out-of-band verification and approval controls. Optional simulated scenarios, and a board-ready report. Advisory readiness, not covert testing or ongoing monitoring.
Scope a readiness assessment →A full governance framework: policy, model-risk classification, owner matrix, vendor controls, and evidence. It is informed by NIST AI RMF. It is designed to support your EU AI Act compliance program.
Scope a framework build →Retained senior leadership for your AI program, on a fraction of a hire. We own the risk register, board reporting, evidence, and operating cadence.
Scope retained leadership →A point-in-time threat model and adversarial test: prompt injection, tool abuse, data leakage. You get severity-ranked findings and remediation. A retained AI security co-pilot is available from $6,000/mo.
See exact figures on the cost page →Copilots and agents taken to production. That means workflow design, data integration, control implementation, evaluation, and a documented production handoff.
Scope an implementation →Open-weight models on your own cloud or VPC. Proprietary models are available when a workload needs them. Your data stays within your boundary.
Scope a private LLM platform →A point-in-time assessment of your data platform. You get a prioritized plan to make the data actually ready for production AI.
Scope a data assessment →Pipelines, data quality, and governance, built and handed back. Your team gets a runbook it can operate.
Scope a data build →Applied data science taken all the way to production. That covers evaluation, ML engineering, and the data foundations. Together they move a model from notebook to a system you own.
Scope an ML build →A fixed-fee, point-in-time read on hardening, MFA, phishing exposure, and backup-restore. Plus a cyber-insurance-questionnaire readiness checklist.
Book a free 30-minute Cyber Risk Check →Pick an engagement type and describe the shape of the work. You get a market-estimate range for that model, drawn from the approved bands above. It is always a range, never a single number. The tool runs entirely in your browser. Nothing you enter is sent anywhere.
Choose an engagement type and scope inputs on the left, then estimate. The result is a range, not a quote.
·
··
These are non-binding market estimate ranges, not quotes. The number narrows, up or down, once we scope the work together and fix a fee in writing.
Last reviewed: 2026-07-03 · Initial release. Ranges are a structured practitioner device that reflects market rates for work of each shape. We re-check them quarterly and confirm them on a call. Accuracy is the point.
Straight answers on how the ranges work and what's cheapest to start with.
Governance work is priced by shape. An AI Readiness or Strategy Sprint starts around $12,000 and runs to $55,000 for complex, multi-entity work. A point-in-time AI Governance Gap Assessment runs $35,000 to $55,000. A full AI Governance Framework build runs $55,000 to $220,000, higher for multi-entity or cross-jurisdiction programs. Retained governance leadership (a fractional AI vCISO) runs $8,000 to $18,000 a month.
Fixed fee, always. Every range on this page is a non-binding market-estimate range, not a quote. We scope the actual work on a call and confirm the fee in writing before anything starts, so you approve a number, not an open-ended meter.
The Security Posture Assessment starts at $1,500 as a one-time, fixed-fee entry point for small businesses. For AI-specific work, the AI Security Assessment / Red Team starts at $6,000, and a retained AI security co-pilot is available from $6,000 a month.
Retained AI governance or security leadership runs $8,000 to $18,000 a month on this page's fractional-vCISO tier, scoped to the number of AI systems and the regulatory surface you face. See the vCISO for AI page for how a narrower AI-program vCISO retainer is scoped starting from $6k/mo.
No. Every number on this page is a non-binding market-estimate range built from the shape of similar past engagements. The number narrows, up or down, once we scope your specific work together and confirm a fee in writing.